U.S. patent application number 13/635723 was filed with the patent office on 2013-08-08 for method for testing hearing aids.
This patent application is currently assigned to SIEMENS MEDICAL INSTRUMENTS PTE. LTD.. The applicant listed for this patent is Martina Bellanova, Matthias Latzel. Invention is credited to Martina Bellanova, Matthias Latzel.
Application Number | 20130202124 13/635723 |
Document ID | / |
Family ID | 43902778 |
Filed Date | 2013-08-08 |
United States Patent
Application |
20130202124 |
Kind Code |
A1 |
Bellanova; Martina ; et
al. |
August 8, 2013 |
METHOD FOR TESTING HEARING AIDS
Abstract
A method for testing hearing aids, particularly in respect of
the effect thereof on speech comprehension, is improved. The method
has the following steps: a) a test system is provided, b) an
audible voice signal in the form of at least one meaningless
syllable is selected and presented c) the meaningless syllable and
a number of further meaningless syllables are displayed on a
graphical user interface in the test system, d) a heard meaningless
syllable is selected from the displayed meaningless syllables by
the user, e) the selection made is evaluated, and f) method steps
c) to e) are repeated until a termination criterion is encountered.
The audible voice signal is selected in each case on the basis of
at least one selection made by the person in response to a
previously presented voice signal. The method allows particularly
fast and informative hearing aid tests.
Inventors: |
Bellanova; Martina;
(Erlangen, DE) ; Latzel; Matthias; (Eggolsheim,
DE) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Bellanova; Martina
Latzel; Matthias |
Erlangen
Eggolsheim |
|
DE
DE |
|
|
Assignee: |
SIEMENS MEDICAL INSTRUMENTS PTE.
LTD.
SINGAPORE
SG
|
Family ID: |
43902778 |
Appl. No.: |
13/635723 |
Filed: |
March 10, 2011 |
PCT Filed: |
March 10, 2011 |
PCT NO: |
PCT/EP11/53600 |
371 Date: |
November 9, 2012 |
Current U.S.
Class: |
381/60 |
Current CPC
Class: |
A61B 5/121 20130101;
H04R 25/70 20130101; H04R 25/30 20130101; A61B 5/7435 20130101;
A61B 5/002 20130101; H04R 2225/43 20130101 |
Class at
Publication: |
381/60 |
International
Class: |
H04R 25/00 20060101
H04R025/00 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 18, 2010 |
DE |
10 2010 011 950.4 |
Claims
1-7. (canceled)
8. A method for testing speech comprehension of a person assisted
by a hearing aid, which comprises the following steps of: a)
providing a test system; b) selecting and presenting an acoustic
voice signal in a form of at least one meaningless syllable; c)
displaying the meaningless syllable and a number of further
meaningless syllables on a graphical user interface of the test
system; d) selecting, via the person, a heard meaningless syllable
from displayed meaningless syllables; e) evaluating a selection
made; and f) repeating the method steps c) to e) until a
termination criterion is reached, the selection of the acoustic
voice signal being made in each instance in dependence on at least
one selection made by the person in response to a previously
presented voice signal.
9. The method according to claim 8, which further comprises
superimposing an interference signal on a presented voice
signal.
10. The method according to claim 9, which further comprises
providing a fluctuating interference as the interference signal to
be superimposed.
11. The method according to claim 10, which further comprises
presenting the acoustic voice signal in a region of a minimum
fluctuating noise interference.
12. The method according to claim 8, which further comprises
performing at least one of a selection or a presentation of at
least one of the acoustic voice signal or of the interference
signal to repeat the method steps c) to f) in dependence on an
evaluation of the selection made by the person in relation to at
least one previous round of the method steps c) to f).
13. The method according to claim 12, which further comprises
adjusting a signal to noise ratio between the acoustic voice signal
and the interference signal.
14. The method according to claim 8, which further comprises
adjusting a degree of similarity between the meaningless syllable
selected and the further meaningless syllables.
15. The method according to claim 9, which further comprises using
noise as the interference signal.
Description
[0001] The invention relates to a method for testing speech
comprehension of a person assisted by a hearing aid.
[0002] Hearing aids can help a person with diminished hearing
perception compared with a person with normal hearing to perceive
his/her environment in the same way as the person with normal
hearing. However this is only possible with a high-quality hearing
aid, which has to be adjusted precisely to the individual hearing
loss of the person in question. It is particularly important for
hearing-impaired people that they are able to converse with other
people and therefore comprehend speech. It is therefore expedient
and necessary to test the quality of a hearing aid and/or the
quality of its adjustment in particular in respect of voice
signals.
[0003] Hearing aids are already tested regularly during the
development stage or after manufacture. Tests are also carried out
in particular to determine the extent to which certain algorithms,
for example for compression, frequency compression, noise
interference suppression, etc., influence the ability of a person
assisted by the hearing aid to comprehend speech. However speech
comprehension tests can also be expediently carried out during or
after hearing aid adjustment.
[0004] A method for testing and optimizing hearing aids is known
from the publication US 2005/0027537 A1, in which a user assisted
by a hearing aid is presented with meaningless syllables. Settable
parameters of the hearing aid are reset as a function of whether
the presented syllables are correctly understood.
[0005] The object of the present invention is to improve a method
for testing hearing aids, in particular in respect of their effect
on speech comprehension.
[0006] This object is achieved by a method with the method steps as
claimed in claim 1.
[0007] A test system is provided to implement the method. The test
system comprises a control unit for controlling the test. A
meaningless word is first selected, preferably from a pool of
stored meaningless words. The selected word is then presented to
the person wearing the hearing aid (test subject) by way of an
output medium, preferably a loudspeaker. A wired or wireless
electrical signal transmission between the test system and the
hearing aid is also possible however.
[0008] When assessing the quality of a hearing aid, the
presentation of voice signals provides more reliable information
compared with the presentation of sinusoidal tones. So that the
hearing aid test can be performed internationally without
modification, particularly language-independently, the voice signal
presented preferably comprises meaningless syllables, referred to
as logatomes. These can consist for example of
vowel-consonant-vowel sequences (aba, aca, ada, . . . ) or
consonant-vowel-consonant sequences (bab, beb, bib, . . . ). In
some instances these meaningless syllables should be checked again
and only those that have no meaning in all the major languages
should be selected. The syllables presented preferably come from
voice recordings, however they can also be synthetically
generated.
[0009] A presented syllable (e.g. aba) is then displayed together
with other similar-sounding syllables (e.g. ada, aga, aha, aka, . .
. ) on a graphical user interface of the test system. The test
subject selects the syllable he/she thinks he/she heard from the
displayed syllables, for example using a pointer. Alternatively the
test subject can repeat the syllable he/she identified and this can
be captured by a speech recognition system.
[0010] The selection made is then evaluated. In the simplest
instance the selection is identified as correct or incorrect.
[0011] The test is repeated with a new selection of meaningless
syllables, until a termination criterion is reached.
[0012] According to the invention the acoustic voice signal is
selected after each round as a function of at least one selection
made by the test subject in response to a previously presented
voice signal. The selection made by the test subject in response to
the voice signal is therefore evaluated immediately after every
round. The last selection and preferably also the results of
previous rounds form part of the selection of a new voice signal,
in particular a logatome. The selection of the voice signals
(logatomes) is therefore adaptive, based on the "previous history".
The adaptive selection of the voice signals from the voice signal
pool constantly allows voice signals to be presented that are most
likely to produce responses to the questions arising in the current
test. Voice signals that are less suitable for the current test are
passed over. This shortens the test time, since the test subject
does not have to go through the entire inventory of voice signals
(signal pool) for every test. The specific selection of voice
signals and the gradual focusing of the voice signals during the
ongoing test will detect and define the problem areas for the test
subject in respect of speech comprehension, e.g. their ability to
identify certain consonants correctly.
[0013] In one preferred embodiment of the invention an interference
signal, in particular noise, is superimposed on the presented voice
signal. When voice signals are presented in noise interference, the
amplification setting of the hearing aid of the test subject has a
secondary role. The test signals then only have to be presented at
one volume, at which they can be clearly perceived. The sensitivity
of the hearing test results from the variation of the signal to
noise ratio SNR. The test starts initially with no or little noise
component and the noise component is then increased progressively
in each round in relation to the voice component, until the test
subject no longer identifies the voice signal correctly.
[0014] The ratio of voice signal to interference signal, at which
the test subject no longer identifies the voice signal, then allows
reliable information to be derived about how good the hearing
capacity of the test subject is with a hearing aid and therefore
the quality of the hearing aid or assistance.
[0015] The reliability of the hearing aid test or the information
it provides can be further improved by using fluctuating noise, in
particular amplitude-modulated noise with a preferably random
modulation, as the interference signal. The use of modulated noise
interference provides a high level of sensitivity in respect of a
hearing reduction, since the threshold for modulated sound is
raised significantly for inner ear hearing problems in particular
compared with normal hearing.
[0016] Particularly reliable and good results are also achieved if
the voice signal is presented in each instance in the region of a
minimum fluctuating interference signal.
[0017] In one embodiment of the invention not only the voice
signals per se but also the response alternatives presented in
response to a voice signal are adjusted adaptively to the test
pattern to date. In this process the degree of similarity of the
presented alternatives to the correct solution can vary. On the one
hand therefore alternatives can be presented which have a high
degree of similarity and on the other hand alternatives can be
presented which have little similarity to the correct solution.
[0018] Like the selection of the voice signals, the selection of
noise interference, the setting of the signal levels and in
particular the level ratio between voice and noise interference and
the timing of the voice signal in fluctuating noise are also
effected adaptively, in other words as a function of the results of
previous test rounds.
[0019] The inventive method can be used for a qualitative and
quantitative determination of the improvement in speech
comprehension as a result of the use of a hearing aid. It is also
possible to perform a comparison between different hearing aids on
the same test subject. Also different settings of the same hearing
aid can be tested for their influence on speech comprehension, for
example the influence of a compression, a frequency compression or
a noise interference elimination algorithm.
[0020] The invention is described in more detail below with
reference to a flow diagram. The FIGURE shows the different
stations for the performance of a hearing aid test according to the
invention.
[0021] First a logatome is selected randomly from a pool of
logatomes 1 and presented by way of a loudspeaker 2 to a test
subject 3 assisted by a hearing aid. The test subject is also shown
the acoustically presented logatome together with similar logatomes
on a display 4 for selection purposes. The test subject selects
from the logatomes presented on the display the logatome he/she
thinks was presented acoustically. The test subject's selection is
first analyzed in an analysis and selection facility 5, in
particular to determine whether the selection was correct.
Depending on the result of the analysis, a new logatome is then
specifically selected and presented to the test subject. This
closes the circle and the method is continued until a termination
criterion is reached.
* * * * *